User talk:CmdrObot

From Wikipedia, the free encyclopedia

Welcome to CmdrObot's talk page.

Please sign your comments using four tildes (~~~~). Place comments that start a new topic at the bottom of the page and give them ==A descriptive header==. If you're new to Wikipedia, please see Welcome to Wikipedia and frequently asked questions.
Archive
Archives
archive 1 (Mar. 18, 2006 – Sep. 2, 2006)
archive 2 (Sep. 2, 2006 – Nov. 7, 2006)
archive 3 (Nov. 7, 2006 – Jan. 22, 2007)
archive 4 (Jan. 23, 2007 – Jul. 8, 2007)
archive 5 (Jul. 26, 2007 – Jan. 8, 2008)

Contents

[edit] Incorrect correction for "institution"

I noticed an incorrect spelling correction to Fort Worth, Texas. diff. The bot changed "instiution" to "istitution" when I believe it should have become "institution".  :)

Thanks for a great bot! --Hebisddave (talk) 17:54, 9 January 2008 (UTC)

Oh dear, that's embarrassing! I've updated the bot, added a rule to correct "istitution" whenever it finds it, and searched wikipedia for any instances of "istitution" in the last database dump. Thanks for letting me know about the error. Cheers, CmdrObot (talk) 20:10, 9 January 2008 (UTC)

[edit] Character names

One possible problem with the bot - names could be considered "mispellings" of the words that the bot "fixes". For example, on one of the recent pages, it edited an octopus-creature's name from "Octogon" to "Octagon". I have no idea how to program it to do so, but you should be able to solve this problem by having the bot avoid capitalized words.Not even Mr. Lister's Koromon survived intact. 01:33, 13 January 2008 (UTC)

Hi KrytenKoro, sorry about that miscorrection. The problem with your suggestion is that you frequently get capitalised words that are misspelled too, so I can't just ignore capitalised words in general. However, I will add an exception for 'Octogon' to prevent it from happening again. Cheers, CmdrObot (talk) 21:20, 13 January 2008 (UTC)

[edit] List of corrections?

Is there a publicly viewable list of the corrections that CmdrObot makes? I've run across some indefinite article mix-ups that irritate me ("an ewe"—though be careful of the Ewe, "an unique", "an unicorn", "an European", and so on), but I don't know whether to suggest them here as I've just done or whether this bot is already taking care of it and will get around to them sooner or later. —Bkell (talk) 06:18, 15 January 2008 (UTC)

Oh, ha, if I would just take the time to look in the archives I see there's a link to User:CmdrObot/SpellList.py. This seems to have only spelling corrections, though, unless I just don't understand how it works. —Bkell (talk) 06:22, 15 January 2008 (UTC)
Hi Brian, thanks for your interest. The bot uses several different strategies for attempting to correct articles. The first one is to perform a word-by-word search and replace on the article using the misspellings in SpellList.py. Another one is to use a list of regular expressions to handle some of the trickier cases that span multiple words or may be correct depending on the surrounding context. Finally, the bot uses a number of ad-hoc routines to deal with things such as 'broken' links (eg ones like http://http://foo.com, or http:////foo.com); external links being used instead of wikilinks; some specific uses of greengrocers' apostrophes and so on. The bot's handling of a/an mismatches is one of these ad-hoc routines.
The routine assumes that 'a' followed by a word beginning with a vowel is incorrect, as is 'an' followed by a word beginning with a consonant. Obviously there are numerous exceptions to this--in English 'eu' words and 'h' words are common examples, as are acronyms such as 'MRI' and 'XML'. Just to make life a little more exciting, 'a' and 'an' play a different role in French, Spanish and German. I get the bot to check against a list of a few hundred of these common exceptions before trying to make any changes, and this keeps the false-positives down to a tolerable level.
Thanks for making me explain this, BTW, because as a result of doing so, I've realised there's a mistake in the bot. It correctly ignores things like 'an XML', but I forgot to handle the converse: it doesn't try to correct 'a XML'. I guess that gives me something to work on this evening :). Cheers, CmdrObot (talk) 14:00, 15 January 2008 (UTC)

[edit] Feature request: Alphabetizing 'see also' sections and 'disambiguation' pages

It would be useful if your bot could automatically alphabetize 'see also' sections and disambiguation pages. Disambiguation pages would be a little more complicated on account of the subheadings - each subheading should probably be in the proper alphabetical order and each item under each subheading should also. Do you think this is doable? ----Seans Potato Business 14:38, 15 January 2008 (UTC)

[edit] An LLC?

Recently the bot made a correction on Articles of Organization. It changed "a LLC" to "an LLC". Although "An El El See" sounds better when spoken, I believe it is still "A LLC" since "L" is not a vowel. -- Emana (Talk) 20:39, 22 January 2008 (UTC)

Hi Emana, thanks for your comment. If you look at the A and an article though, you can see that current English usage is to use "an" when the following noun begins with a vowel sound. Cheers, CmdrObot (talk) 21:40, 22 January 2008 (UTC)
Thanks for the reply. I've done some digging and found out about the phonetic usage. I never knew about the "Y" sound rule either. I think this is the first really useful stuff I've learned on WP. -- Emana (Talk) 22:26, 22 January 2008 (UTC)
But I don't think the "Y" rule should apply in this case where the bot changed it from "an unanimous"→to "a unanimous". After all it's not ynanimous is it? But for clarity I rewrote the sentence. 08:43, 29 February 2008 (UTC) —Preceding unsigned comment added by 66.39.171.19 (talk)

[edit] "A NES" versus "An NES"

Regarding this edit, it is well-supported that the acronym can be pronounced "en-ee-es", "ness", "nezz", and even perhaps "Nintendo"; there is certainly no consensus on the issue. The same situation arises with SNES (where it is even mentioned and sourced); my solution for that article was to avoid using the indefinite article completely in connection with the abbreviation. Anomie⚔ 01:31, 25 January 2008 (UTC)

Hi Anomie, thanks for that. I'll get my bot to refrain from modifying SNES or NES in future. Cheers, CmdrObot (talk) 13:22, 25 January 2008 (UTC)

[edit] An urinalysis

See [1]. It doesn't sound right to me, so I'll change it back. Not sure if the bot requires a tweak or if this was an isolated incident (or if I've just been saying it wrong these many years). --Joelmills (talk) 03:51, 25 January 2008 (UTC)

Hi Joel, thanks for that. No, it's not you, it's me: I slipped up on that article. I've updated the bot so it won't try to correct that again. Cheers, CmdrObot (talk) 13:20, 25 January 2008 (UTC)

[edit] 'An RAAF'

This bot is replacing mentions of 'a RAAF' [Royal Australian Air Force] to 'an RAAF' (as one example: [2]. This is not grammatically correct - can you please stop it doing this? Thanks, --Nick Dowling (talk) 02:50, 26 January 2008 (UTC)

Hi Nick. Could you tell me how this isn't correct? If you look at A and an it talks about how an and a are used in English. The acronym is pronounced 'Or aye aye eff' right?
Cheers, CmdrObot (talk) 02:57, 26 January 2008 (UTC)
The acronym is normally pronounced 'r double aye eff' or 'raff'. I've read an awful lot of Australian military history and have never seen a reference to 'an RAAF' thing. --Nick Dowling (talk) 05:22, 26 January 2008 (UTC)
Hi Nick, since it's usually written 'a RAAF', I'll defer to convention and get my bot to leave it alone. You've got to admit that it doesn't quite sound right though. In computer networking, there's a set of standards documents called RFCs (Requests For Comment), and one of those documents will be referred to as 'an RFC', rather than 'a RFC' because in the latter there's an awkward stop between the 'A' and the 'R', whereas 'An R' rolls off the tongue more easily. Anyhow, thanks for the feedback, and please let me know if the bot causes you any other problems. Cheers, CmdrObot (talk) 19:38, 26 January 2008 (UTC)
It's the Australian accent. Clerks. (talk) 21:40, 30 January 2008 (UTC)
As Nick said, it's usually an acronym pronounced "raff", though often an initialism pronounced "are-double-ay-eff".
So, in conversation you hear both "it's a raff base" and "that's an are-double-ay-eff plane". (ay as in hay)
However, as Nick said, when written, I have only ever seen "a RAAF".
(I can't explain why, and now I've read the above, I will forever be uneasy about it!!) Pdfpdf (talk) 00:49, 2 February 2008 (UTC)
Rereading the above, I can see that the explanation is "a Royal". Pdfpdf (talk) 01:06, 2 February 2008 (UTC)

[edit] correcting refs

Regarding a recent edit to Grey-capped Warbler, CmdrObot corrected a sentence which was possibly misspelled (sp: an unique→a unique). While this may be correct (I always thought that an was used for words starting with vowels, interesting that I am wrong, must read about that) the correction was to a reference and the article cited had the error in the title, which was copy-pasted direct. I think our cites should match the cited work accurately, could CmdrObot flag changes to references sections for people who follow the articles to check the changes maybe? I'm guessing most mistakes in citations are editor errors here rather than in the initial cited work, and that this isn't massively important, but I thought I'd mention it. Sabine's Sunbird talk 01:34, 29 January 2008 (UTC)

Hi Sabine, thanks for the feedback. You're completely right, I should be leaving references alone. I've added an exception to my bot so it won't try to miscorrect that title again. Your suggestion of flagging reference sections is an interesting one. I'll add it to my to do list :) Cheers, CmdrObot (talk) 01:44, 29 January 2008 (UTC)

[edit] peer review

I was wondering if you could take some time out of your schedule to head over to the Heroes (TV series) talkpage and give us an honest peer review. The page has gone through some major changes in the last few months, and it would be fantastic if a prominent editor/contributor like yourself, could head over and give us at the Heroes Wikiproject some sound opinion and ideas on improvements for the page. We have all worked very hard at improving the page, and we need great outside, reliable and trustworthy users to come over and help us improve. I you are interested in joining the peer review discussion with other prominent users/contributors, much like yourself, please follow the link. Thank you very much for your help and your continued effort to improve Wikipedia and its quality! Wikipedia:Peer review/Heroes (TV series)/archive2--Chrisisinchrist (talk) 05:58, 27 January 2008 (UTC)

[edit] an, again

Surely replacing "a Sc D" by "an Sc D" as the bot did on George G. Hall is wrong. --Bduke (talk) 02:30, 31 January 2008 (UTC)

Hi Brian. "Sc. D" is pronounced "ess cee, dee" right? If that's the case then going by the "use 'an' before vowel sounds" rule on A, an is the correct thing to do. Unless, of course, I'm missing something, which is always a possibility. Cheers, CmdrObot (talk) 02:39, 31 January 2008 (UTC)
Maybe. I do not pronounce abbreviations when I read them, so "a" rather "an" seems natural to me. It would be "a" if the abbreviation was spelled out. --Bduke (talk) 03:55, 31 January 2008 (UTC)

[edit] The article for usurper is "a", not "an".

This was incorrectly changed in article Dragon Dance (novel).

Hi Snowybeagle. According to Google it could go either way (~80,000 hits for 'a usurper' vs ~30,000 hits for 'an usurper), so I'll get my bot to leave either version alone. BTW, could you sign your posts in future? Thanks, CmdrObot (talk) 01:07, 1 February 2008 (UTC)

[edit] Changes inside quotes?

Three questions: Your bot changed "U.K." to "UK"

  1. I was wondering why? (i.e. What's wrong with U.K.?)
  2. Of more interest to me: In this case, the "U.K." was inside a piece of quoted text, and hence, presumably, should have been left alone? (i.e. What's your POV for this sort of circumstance?)
  3. What about text inside <!-- -->?

Thanks, Pdfpdf (talk) 01:00, 2 February 2008 (UTC)

Hi PDF, yes I probably should have left stuff inside quoted text alone. Could you tell me what article the bot edited so I can add it to my exception list? I added the U.K.→UK rule to my bot after this exchange with another user. Cheers, CmdrObot (talk) 16:22, 2 February 2008 (UTC)
You definitely need to check that you are not altering quotations, as you did in spectral evidence. This should be addressed not through an exception list, but by checking whether the material is within quotation marks, for this will occur in hundreds of articles, and your changes may go uncorrected. - Nunh-huh 22:22, 3 February 2008 (UTC)
Hi Nunh-huh, I agree with you, but unfortunately this isn't nearly as simple to do as it might sound. To see some of the problems involved, have a look here]. Any thoughts? Cheers, CmdrObot (talk) 22:35, 3 February 2008 (UTC)
I'm not certain what the programming difficulty is; it seems that double quotation marks are not part of Wiki markup, and so it should be possible to simply count quotation marks from the beginning of an article and know if you are in quoted text or not (odd number of quotation marks = you're in quoted text). And of course you'd have to allow for <blockquote></blockquote> as well. Not that I'd be surprised if I'm overlooking something. But I do think it's far, far better to leave errors uncorrected than to introduce new ones by "correcting" accurate quotations. - Nunh-huh 00:03, 4 February 2008 (UTC)

Sorry CmdrObot, but I'm afraid I am of a similar opinion to Nunh-huh. (i.e. it's far, far better to leave errors uncorrected than to introduce new ones)
Also, what about text inside <!-- -->?
And I thought of another situation: Citations. (e.g. If a citation says "U.K.", is that the same as a quotation, and should not be changed?)
Cheers, Pdfpdf (talk) 11:22, 4 February 2008 (UTC)

Trust me, it's not simple to handle the general case, it really isn't. However, I have made some modifications to the bot that should cover some of the common cases. The bot now ignores
  • Text inside a <blockquote> tag
  • Text inside one of the 25 or so templates used explicitly for quoting (eg Template:cquote)
  • Plain text with no wiki markup inside double quotes.
Running this over a couple of thousand articles from a database dump, I see that the bot now ignores a few quoted corrections that it should have before, but also incorrectly misses a few things that it should have corrected. I can live with that tradeoff for now--it's not like wikipedia lacks for things that need to be corrected. Cheers, CmdrObot (talk) 19:45, 6 February 2008 (UTC)

[edit] RSTS

Hi... Your bot changed "a RSTS/E user..." to "an RSTS/E user...". But "RSTS" is pronounced as if it were a word, not letter by letter, so the previous text was correct. Could you add this to your exceptions? Thanks. Paul Koning (talk) 22:30, 6 February 2008 (UTC)

Hi Paul. Sorry about that. Exception duly noted! Cheers, CmdrObot (talk) 22:56, 6 February 2008 (UTC)

[edit] a MK -> an MK

Hello! I noticed in this edit that you changed "a MK 14 Mod 3 warshot torpedo" to "an MK 14 Mod 3 warshot torpedo". MK is an abbreviation for "mark", and generally if the above phrase were being read out loud you'd say "mark" instead of "em-kay", so "a MK 14" is preferable. TomTheHand (talk) 21:08, 11 February 2008 (UTC)

Hi Tom, thanks for letting me know. I've added it to my bot's exception list so it won't do that anymore. Cheers, CmdrObot (talk) 21:10, 11 February 2008 (UTC)

[edit] a/an AAA

Hi,

On this edit the bot replaced "a AAA" with "an AAA." Since "AAA" is pronounced "Triple-A", the preceeding "a" is correct. I thought I'd point this out so it could be fixed. -NatureBoyMD (talk) 21:55, 11 February 2008 (UTC)

Hi there. Thanks for letting me know. I've updated my bot's exception list to take it into account. Cheers, CmdrObot (talk) 18:18, 12 February 2008 (UTC)

[edit] SAS

In the SAS System and SAS Institute articles. SAS is historically pronounced "sass", so "a SAS" is correct, not "an SAS". Cheers, Metamusing (talk) 16:31, 22 February 2008 (UTC)

Hi Meta, thanks for that. I've added SAS to my list of exceptions, sorry for the inconvenience. CmdrObot (talk)

[edit] Change only clear cases? - bot suggestion

How about continuing to change articles before acronyms if the same article would be used when the acronym was treated as a word. When a different article would be used if the acronym was treated as a word, perhaps generate a list to be manually checked before automatically changed. This might increase the ratio of work saved to work generated for other editors. Thanks and Cheers, Metamusing (talk) 16:40, 22 February 2008 (UTC)

Hi Meta. I'm not quite sure what you mean by 'treated as a word'. Do you mean check the acronym as if it were all lower case letters first?
I think it might not be worth doing; the reason I've got a number of complaints so far is that this is my first sweep over wikipedia with the bot with acronym-checking code, so I've had to build up its exception list to 'teach' it all the cases I should leave alone. That sweep is almost finished now (and I've a big long exception list, most of it populated by things I could figure out for myself, but also with help from other editors like you for the things I didn't know). Also, after the sweep is done, I'll go back to my ordinary spell checking rather than focussing on a/an, so the number of time I come across a/an, let alone get it wrong, will go down quite substantially.
Hope this helps. CmdrObot (talk) 23:42, 22 February 2008 (UTC)
Right, as if it were lower case letters. Thank you for your consideration. I'm glad you're trying to figure out for yourself if corrections are appropriate. Cheers, Metamusing (talk) 20:10, 23 February 2008 (UTC)

[edit] It would be better not to interfere with correct historical usage

The bot changed "an universal dictionary ..." in a quoted (and correct) title dating from 1704, to "a universal ...." (Lexicon technicum). (This one has been changed back along with other revisions.) It really would be better not to do that! Terry0051 (talk) 14:43, 23 February 2008 (UTC)

Terry, thanks for that. I've added the spelling to my exception list for that article. Cheers, CmdrObot (talk) 18:31, 23 February 2008 (UTC)

[edit] Correcting capitlisation

Would you be willing/able to have your bot adjust capitalisation of certain phrases? A common problem that seems to be sweeping Wikipedia at the moment is inappropriate capitlisation of the full-form of phrases that also have frequently-used acronyms. Take for example, suppressor of cytokine signaling proteins (SOCS); people assume that because SOCS is written in capitals, the first letter of each word of the full-form is also written in uppercase. Perhaps you could create a subpage for people to add to a list of suitable phrases for automatic de-capitalisation? I know certain items wouldn't qualify, since they might be likely to occur in capitlised-form in legitimate situations, but there are many cases such as the example above that would be apt. Should I start compiling a list? Thanks, ----Seans Potato Business 01:07, 24 February 2008 (UTC)

Sean, that sounds like a good idea. My bot can handle that no problem. I've created a page where you can add phrases. If you could sign each item you add, that would be good. Cheers, CmdrObot (talk) 12:58, 24 February 2008 (UTC)
Thanks a lot; this is really helpful. Regarding the list; should I put extra line breaks in so that the entries don't get bunched up? I added "death inducing signalling complex", because people seem to forget to use the hyphen. ----Seans Potato Business 23:29, 24 February 2008 (UTC)

[edit] Fuze is the usual English language spelling for ordnance igniters

Somebody keeps running a robot to change all occurrences of Fuze to Fuse in Shell (projectile). Fuze is the customary English language spelling for ammunition igniters. Fuses are something in an electrical layout.Rcbutcher (talk) 11:14, 1 March 2008 (UTC)

Hi there, my apologies. I had looked this one up in a dictionary, and it didn't indicate the distinction; I guess it just wouldn't have that sort of specialised knowledge. In any case, I've removed the fuze->fuse entry from my bot. Cheers, CmdrObot (talk) 16:02, 2 March 2008 (UTC)

[edit] Analog vs Analogue

The bot just changed it in the Microsoft SideWinder article. According to the Oxford English Dictionary analog is the US variant of analogue and it is correct. —Preceding unsigned comment added by 87.205.248.87 (talk) 12:13, 1 March 2008 (UTC)

I looked at the edit history, and no such change was made (which is not surprising given my bot doesn't have an 'analog->analogue' rule for the reasons given above). It did change "an USB" to "a USB", "a an USB" to "a USB", and "a endless" to "an endless". CmdrObot (talk) 16:09, 2 March 2008 (UTC)

[edit] an MW

In Control of the National Grid (UK), "a MW" was changed to "an MW". SI units would be pronounced in full, so it should be "a MW" --ascorbic (talk) 21:48, 1 March 2008 (UTC)

Thanks Ascorbic, it's a good point. I've updated my bot to deal with this. Cheers, CmdrObot (talk) 16:12, 2 March 2008 (UTC)

[edit] Incorrect Correction

On List of Meerkat Manor meerkats, the bot replaced all instances of Commandoes with Commandos, however the groups name in the series is Commandoes. :) Collectonian (talk) 01:03, 3 March 2008 (UTC)

Oops. Sorry about that. I've added that spelling to the exception list for that article in my bot. Thanks for letting me know. CmdrObot (talk) 14:15, 4 March 2008 (UTC)

[edit] Revert of reference "title" on Orion (spacecraft) page

I've reverted CmdrObot's "correction" of "Nasa" to "NASA". While I agree that "NASA" is correct, "Nasa" is how it is in the title of the BBC News article that the reference cites and should therefore read that way.

However, given how the "Brits" don't seem to know how to spell acronyms in the English language properly, I would be perfectly happy to see "BBC" replaced with "Bbc" in all articles...

-)

Beachgrinch (talk) 01:31, 3 March 2008 (UTC)

Hi Beachgrinch. I've added an exception to my bot fot this, thanks for letting me know. It's an odd error, and I'd have expected better from the BBC — usually they're fairly good about this sort of thing. Cheers, CmdrObot (talk) 14:20, 4 March 2008 (UTC)

[edit] Error

The bot is adding article deletion tags to images [3] [4]. Icestorm815 • Talk 20:57, 5 March 2008 (UTC)

Hi Icestorm. The bot isn't adding deletion tags, I'm doing that by hand myself as I run across nonsense/non-noteable images while using my bot. So far maybe 10 images I've nominated for speedy deletion in the last day have been removed. Do you think the images you mentioned were incorrectly tagged? Cheers, CmdrObot (talk) 21:40, 5 March 2008 (UTC)
Well, I'm not the greatest with images, but I know for a fact that {{db-bio}} is strictly for articles. Images released into the public domain have to be deleted through a deletion discussion. Hope that helps! Icestorm815 • Talk 21:48, 5 March 2008 (UTC)

[edit] Incorrect change to joint ITU-T | ISO/IEC standards references

Your bot is incorrectly deleting the space after the vertical bar in the references to joint standards in the External links section of the X.400 page (and I guess perhaps elsewhere as well). Can you make it revert this incorrect change? Or should I manually revert all its changes? 213.123.172.26 (talk) 05:26, 22 March 2008 (UTC)

[edit] Dubious change to an indefinite article before an acronym

Your bot is making a dubious assumption about how a reader reads an acronym, pronouncing each letter individually. Unless you are familiar with the subject matter you don't know how the acronym is conventionally spoken. Often an acronym is spoken either as if it was a single word sometimes with implied vowels added to make it pronouncable, or as the full set of words it abbreviates. In either of these latter cases the changes your bot is making to the indefinite article are incorrect.

For example, NASA can be read as a word spoken 'nassah', as letters spoken 'en ay ess ay', or as 'national aeronautical and space administration'.

I think it would be much safer to assume that subject matter editors know what is appropriate rather than have a bot that thinks it knows better.

213.123.172.26 (talk) 05:55, 22 March 2008 (UTC)

[edit] Doroshenko

Now, this couldn't possibly be correct :) I think some algorithm tweaking is in order...—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); 22:06, 2 April 2008 (UTC)

Ah! Tricky. As it happens I had an exception in my bot for "Ukraine" and "Ukrainian" (which use 'a' rather than 'an'), but I didn't have one for the misspelt 'Ukrainan'! I have added a spelling correction rule to my bot for 'Ukrainan' in case this happens again though. Cheers, CmdrObot (talk) 22:12, 2 April 2008 (UTC)

[edit] Correction to Mutants (collection)

Hi - Your bot changed Solvant→Solvent [5] in the article Mutants (collection). I've double checked both of the sources listed in the article and they both list the spelling as "Solvant". Since it is a title to a short story, I suspect the author was trying to make a pun and it may not be an actual word in English. I'm going to change it back (I won't revert since the other correction is good), but I worry that your bot will attempt to re-correct the word "Solvant". The title is already in quotes. Is there a comment or something that could be added to ensure that the bot will not change it back? Thanks.--Rtrace (talk) 02:37, 3 April 2008 (UTC)

Oh dear, my apologies. I've added an exception to my bot so it won't try to 'correct' Solvant in that article again. Thanks for letting me know. Cheers, CmdrObot (talk) 12:07, 3 April 2008 (UTC)

[edit] Open Evangelical

Hi Bot,

You correctly spotted a typo but miscorrected

"which contains an the outline of Open Evangelicalism"

to

"which contains and the outline of Open Evangelicalism"

it should read

"which contains an outline of Open Evangelicalism"

I still think you are clever.

Regards

Springnuts (talk) 06:57, 4 April 2008 (UTC)

Thanks for that, Springnuts. I guess my human supervisor gets distracted on occasion. Such unreliable things, these meat machines! :) Cheers, CmdrObot (talk) 13:20, 4 April 2008 (UTC)

[edit] Keep ENCHANTED DREAMZ page up

Can you please help keep the enchanted dreamz page up, edit it so that i can stay. —Preceding unsigned comment added by Arrowcargo (talk • contribs) 14:41, 4 April 2008 (UTC)

[edit] Rate on quality scale

Hi, I am working on Taiwanese opera and trying to expand it.I saw you are also interested about this topic.I would like to ask you to watch that article and rate on the scope of Wikiproject Taiwan or give me some opinions or comments on this article. ThanksQwaszxfish 04:39, 8 April 2008 (UTC)

[edit] A u* vs. an u*

CmdrObot recently made such a subtitution in Quest for the Well of Souls, converting "A Ulik" to "An Ulik". This is generally an unsafe substitution, as a leading "u" may be pronounced like "yoo" instead of like "oo", and in the former case, the "a" article rather than the "an" article is appropriate. --Mr Wednesday (talk) 08:00, 9 April 2008 (UTC)

Hi Mr Wednesday. Thanks for that. My bot has a big list of exceptions for various different u* words. I'll add Ulik to the 'may be preceded with either a or an' list. Cheers, CmdrObot (talk) 18:09, 9 April 2008 (UTC)

[edit] Micheal Luck

CmdrObot recently changed the spelling of Micheal Luck on his page to Michael. Can this page go on an exclusion list? Regards, Mattlore (talk) 22:02, 13 May 2008 (UTC)

Sorry about that Mattlore. It looks like Micheal->Michael isn't a safe correction in general, so I've removed it outright from my bot. CmdrObot (talk) 23:12, 13 May 2008 (UTC)

[edit] Other people named "Micheal"

Cmdrbot is erroneously correcting the spelling of some other people who are named Micheal, not Michael, including Micheal Williams and Micheal R. Williams. Please don't do this as a universal correction. --Orlady (talk) 22:23, 13 May 2008 (UTC)

I've also undone your bot's corrections in Micheal Nakamura‎ and Micheal Spurlock‎. I can't tell if Micheal Baldwin needs to be reverted, as the referencing is sparse, but the only cited sources also spell the name "Micheal." I expect that there are plenty more people who truly spell the name that way... --Orlady (talk) 22:35, 13 May 2008 (UTC)
My apologies for that Orlady, and thankyou for reverting those mistaken edits. I've removed that rule from my bot (and added a unit test that will complain if I absentmindedly try to add it again in the future). CmdrObot (talk) 23:15, 13 May 2008 (UTC)
Similarly on the page Interstate 70 in Utah the bot substituted Mathew to Matthew. According to the source cited, Mathew is correct for this person. Considering that "alternate" spellings of names seem to be more and more common with each new year, perhaps the bot should not try to correct names, unless unless many spelling errors are present and it is safe to assume the editors are not good spellers.Dave (talk) 02:54, 15 May 2008 (UTC)

[edit] Mike Hudema

The correct spelling of Mike Hudema's first name is in fact with the E before the A, so please add that to your exclusions list. Sarcasticidealist (talk) 07:54, 16 May 2008 (UTC)

[edit] Incorrect correction on Cumbria

The bot changed "Cumbria is one of the countrys least ethnically diverse counties" to "Cumbria is one of the countries least ethnically diverse counties" when it should have been changed to "Cumbria is one of the country's least ethnically diverse counties" --Kayedj (talk) 11:12, 16 May 2008 (UTC)

Ah oops. Thanks for catching that. Cheers, CmdrObot (talk) 18:50, 16 May 2008 (UTC)

[edit] Alexander Anderson (poet)

Your bot 12:41, 17 May 2008 CmdrObot (Talk | contribs) m (3,819 bytes) (sp: mens→men's) (undo) changed "mens divinor" in the Alexander Anderson (poet) article to "men's divinor". But the original was correct. It's Latin. I think it means "a mind more divine", or something like that, although the phrase doesn't appear to be in common use now. Perhaps it should have been italicised. SamuelTheGhost (talk) 13:29, 17 May 2008 (UTC)

Sam, thanks for that. I try to watch out for Latin phrases like that (most recently I saw one in Catullus 68 for example), but I guess sometimes one must slip through the net. I've now added the Anderson one to my exception list. Cheers, CmdrObot (talk) 13:51, 17 May 2008 (UTC)

[edit] Gentlemen's and Ladies

Why in Skiff Championships Regatta did you change Gentlemens to Gentlemen's but did not change Ladies to Ladies' ? If you bot isn't clever enough to sort out the context, ought it to leave well alone to avoid creating such inconsistencies? David Biddulph (talk) 08:22, 18 May 2008 (UTC)

Hi David. Thanks for the pointer, but don't you think you're being a bit on the harsh side? The bot did that substitution because 'gentlemens' is always incorrect for one reason or other, but 'ladies' is usually correct. While you could well argue that I should have noticed the inconsistency being introduced before approving the edit, you could equally well argue that the change that was made was at least a step in the right direction. Cheers, CmdrObot (talk) 18:48, 18 May 2008 (UTC)

[edit] Preformed v performed

Just wanted to let you know the bot correction of "preformed" on the degranulation page was inaccurate - preformed, in that case, meant formed beforehand and wasn't a typo. I don't know if this may come up on other pages, but you may need to keep a look out. ~ Ciar ~ (Talk to me!) 19:40, 25 May 2008 (UTC)

D'oh! Sorry about that. I have been keeping an eye out for that usage, but obviously not a close enough one. I've added the degranulation case to my bot's exception list so it won't happen again. Thanks, CmdrObot (talk) 19:45, 25 May 2008 (UTC)
Also in this edit. I don't think it's safe to assume that "performed" is always the right spelling, and would suggest that this edit at least ought to be manually approved in each instance. Smith609 Talk 12:09, 26 May 2008 (UTC)
Hi Smith, thanks for pointing that one out. I've also added it to my exception list (along with exceptions for 'preformed' in 86 other articles there already). For what it's worth, I totally agree that preformed->performed is not the sort of substitution that can safely be done automatically, but I don't do any fully automatic editing with this bot. Every change it does has to be manually approved, it's just that occasionally I slip up when reading the diffs and an incorrect edit gets through. Anyhow, thanks again. CmdrObot (talk) 13:23, 26 May 2008 (UTC)
Great, thanks for clearing that up! Smith609 Talk 09:41, 29 May 2008 (UTC)

[edit] Bot Code?

Is the bot code available? I'd like to use it on one of my wikis, if that'd be OK by you. If not, please let me know either on this talk page or via e-mail. Regards, -- Joe Beaudoin Jr. Mail Me 23:31, 25 May 2008 (UTC)

[edit] Bete Grise Light‎

Noticed what you just did with some broken URLs on Seul Choix Point Light. This one has some too. Thanks. 7&6=thirteen (talk) 20:21, 29 May 2008 (UTC) Stan

[edit] Amazon links

Hi! A suggestion: instead of remedying Amazon links, wouldn't it be better to gather the ISBN-10 code from the Amazon page (or from the URL itself, if identifiable), then remove the link and add the ISBN magic word after the (now unlinked) text, between parenthesis? After all, this way people can have a fair choice of sources for the books they're interested enough in to actually click the link. Beside, those (like me) who do prefer Amazon, can always install the User:Lunchboxhero/monobook.js script, which is referred to whenever one opens the ISBN-linked page. -- alexgieg (talk) 01:27, 30 May 2008 (UTC)

Hi Alex, that's not a half bad idea. It also has the advantage of increasing the 'semanticness' of the markup. I'll have a look at implementing this when I get some free time. Cheers, CmdrObot (talk) 15:15, 1 June 2008 (UTC)

[edit] Typographic examples mangled -- incorrectly modifies Spacing/Punctuation

Plenken and Klempen had their examples re-formatted into Wikipedia's preferred format, thereby rendering them useless (now fixed). SmackBot had the same problem with French spacing's Unicode example, which Rich Farmbrough put a temporary fix in for (~2008.03.31).

I'd rather not exclude CmdrObot from these articles.

Is it possible, perhaps, for CmdrObot to NOT make stylistic tweaks to articles in the Typography category? Also, perhaps Punctuation and Spacing. Which, I just noticed, Plenken and Klempen should also be tagged as...

cheers, Saltation (talk) 17:21, 31 May 2008 (UTC)

Hi Saltation. Firstly, my apologies for messing up those articles on you. It must be annoying to repeatedly have to fix up this kind of edit. I like your idea of turning off unicode conversion and mucking about with spaces before/after punctuation for articles in specific categories, and I'll have a go at adding this to my bot shortly. In the meantime, I've updated my list of 'entities that are dangerous to automatically convert to UTF-8' to include everything from &#8192;-&#8201; and &#8239;. Do you think this will help? Cheers, CmdrObot (talk) 15:32, 1 June 2008 (UTC)
Hi CmdrObot. Firstly: pish. "Truth emerges more readily from error than from confusion." (Sir Francis Bacon) A minor oversight, easily corrected, made by someone genuinely striving to improve a much much larger result, is a price I'd pay a hundred times over without complaint. This is one very peculiar very specific subset of normal usage, and there's no way any normal person could be expected to know of it. With an active community and effortless regression, we are all your million eyes finding specific/exception(al) instances where your bot's general case can be improved by not being applied.
Secondly, bugfix-wise: interim workaround-wise: I'd suggest #8192-#8205, #8232-#8233, #8239, #8287-#8288, and #8291. Possibly also #8289-#8290? (#8287-#8291 couldn't hurt, in any case)
cheers and thanks. Saltation (talk) 23:10, 5 June 2008 (UTC)
Speaking of obscure character-encoding oversights, use Flock to go to [6], scroll down to 8238, and feel a disbelieving hilarity spread a smile across your face.
And on the topic of Typography and Spaces, have a read of that French spacing article, and after a period of "how geeky is this?" gradually realise that a lot of what you now regard as normal is merely the result of early-days quick-hack computer design and a particular subgroup's fashion sense.
In particular, look at the example of traditional spacing, and consider that, using this format, the standard newspaper font was 6-point, yet fully readable...
Saltation (talk) 23:10, 5 June 2008 (UTC)

[edit] Need spellcheck help

Hi. When you have time, could you please run a spell check on Hoysala Empire and Western Chalukya Empire. Both are my FA's and I have recently added significant number of citations to improve the articles. Later, after a week. please also run spell check on Chalukya dynasty, another FA of mine, which I am currently improving. Thanks, Dineshkannambadi (talk) 18:49, 3 June 2008 (UTC)

[edit] U.K.

In the article on UKZ, the bot incorrectly changed "U.K." (here, a reference to a specific band who use that form of punctuation and not to the country) to "UK". Bondegezou (talk) 08:49, 10 June 2008 (UTC)